Sample Selection Effects Research Articles

Effects of Sample Selection on Estimates of Economic Impacts of Outdoor Recreation

Estimates of the economic impacts of recreation often come from spending data provided by a self-selected subset of a random sample of site visitors. The subset is frequently less than half the onsite sample. Biased vectors of per trip spending and impact estimates can result if self-selection is related to spending patterns, and proper corrective procedures are not employed. This paper shows a method for accounting for both sample selection and the censored nature of reported expenditures, via a Tobit model with sample selection. Results from a sample of visitors to Cumberland Island National Seashore indicate a naive (uncorrected) approach overestimates per trip visitor spending by 15 percent and economic impacts to industrial output by 10 percent.

The Dependence of Galaxy Clustering on Luminosity

We present statistical results of a two-point galaxy-galaxy cross-correlation function analysis. These studies use appropriate cuts in redshift and limiting apparent magnitude in the two catalogs explored and provide useful information about the dependence of galaxy luminosity on local density. We estimate projected two-point cross-correlation functions w(rp) for samples of center galaxies taken from the Las Campanas Redshift Survey with different mean luminosities against Edinburgh/Durham Southern Catalogue L L* galaxies. Power-law best fits for w(rp) with center galaxies L > L* and L < L* give spatial cross-correlation functions with parameters r0 5.5 ± 0.2 h-1 Mpc, γ 1.8 ± 0.2 and r0 3.5 ± 0.6 h-1 Mpc, γ 2.2 ± 0.2, respectively. Possible differences among results of studies of clustering dependence on luminosity reported in previous works might be dominated by systematics because of sample selection effects. Our analyses indicate that the statistical properties of the large-scale distribution of galaxies depend on intrinsic luminosity, arguing against a universal value of the correlation length. Our results suggest a continuous clustering dependence on absolute magnitude, giving support to previous indications that a smaller biasing factor corresponds to low-luminosity galaxies.

Internal Kinematics of Field Galaxies at Intermediate Redshifts

This paper describes the results of a survey of the internal kinematics of intermediate-redshift (z = 0.25-0.45) field galaxies. Spatially resolved spectra of the [O II] ??3726, 3729 doublet emission line have been obtained for 22 galaxies at the Canada-France-Hawaii Telescope. High spatial resolution has made it possible to extract Vrot sin i and [O II] disk scale lengths from each galaxy spectrum using synthetic galaxy rotation curve fitting. About 25% of the galaxies in the sample have [O II] kinematics unrelated to rotation. [O II] emission is concentrated in the nucleus in these kinematically anomalous galaxies. A Doppler ellipse, similar to those found in local dwarf irregular galaxies, has been observed in a z = 0.35 galaxy. An intermediate-redshift Tully-Fisher (TF) relation, defined by 12 kinematically normal galaxies, shows that these galaxies have a systematically lower rotation velocity (i.e., mass) for their luminosity than expected from the local TF relation. These galaxies would have to fade by ~1.5-2 mag to lie on the local TF relation. This is consistent with starbursting dwarf galaxy models. Although the sample is small, there is also a hint that massive galaxies are not as displaced from the local TF relation as low-mass ones. However, as shown using a large sample of local galaxies, the scatter in the local TF relation is large, especially for late-type galaxies. Sample selection effects, particularly in [O II] emission strength, could be responsible for part of the observed TF evolutionary shift if star formation rates were responsible for the local TF relation scatter. A comparison with other works indicates that a mass-dependent luminosity evolution scenario can explain almost all of the available internal kinematics and surface brightness data.

The soft X-ray properties of a complete sample of radio sources

We present the soft X-ray (0.1--2.4 keV) properties of a complete sample of 88 southern radio sources derived from the Wall \& Peacock 2-Jy sample. It comprises 68 radio galaxies, 18 quasars and 2 BL Lac objects. Whereas both BL Lac objects and all but one quasar are detected in the {\it ROSAT} All-Sky Survey, the fraction of detected radio galaxies is only $\sim$ 60 per cent. For the undetected sources upper limits to the X-ray flux are given. We confirm the correlation of the soft X-ray luminosity (\lx) with the core radio luminosity (\lc) for galaxies as well as for quasars using partial correlation analysis, whereas the corresponding correlations between \lx and \lt are probably spurious due to sample selection effects. We also find strong correlations between \lx and \lc for both Fanaroff-Riley type I (FR~I) and type II galaxies. The broad-line radio galaxies (BLRGs) and the quasars are at the top end of the X-ray luminosity distribution and the detection rate of these objects generally is higher than that of the narrow- or weak-lined radio galaxies. This indicates the presence of an anisotropic X-ray component in BLRGs and quasars, as predicted by unified schemes for radio sources.

Soft X-Ray Properties of a Complete Sample of Radio Sources

We investigate the X–ray properties of a complete sample of 88 radio sources derived from the Wall &amp; Peacock 2–Jy sample. We find that Lx correlates well with core radio luminosity for all object classes, whereas the Lx – Ltotal is probably introduced by sample selection effects. Further, evidence for an anisotropic X–ray component in broad line radio galaxies is reported. A full description of the results will be given elsewhere (Siebert et al. 1996).

Radio and X-ray beaming in steep-spectrum quasars

Soft X-ray fluxes have been drawn from the ROSAT All Sky Survey for an unbiased sample of low-frequency-selected radio quasars, the Molonglo Quasar Sample. In agreement with previous studies a close relationship is found between the radio-core and X-ray luminosities, with steep-spectrum quasars following a shallower correlation slope than flat-spectrum objects. Using the core-to-lobe flux density ratio, R, as an orientation indicator, we can study further the aspect dependence of the X-ray emission components in steep-spectrum quasars. In this way, we trace the presence of a highly anisotropic X-ray component, related to the radio core and consistent with Doppler-boosted models. The orientation dependence of this beamed component, together with sample selection effects, can explain fully the different behaviour of quasars in the radio-core-X-ray luminosity plane, which is found to be a strong function of R. A continuous variation of the relative strength of the beamed component with R is in agreement with the unified schemes which describe steep- and flat-spectrum quasars as members of the same population differing only in orientation.

THE EFFECTS OF SAMPLE RESTRICTION ON GENDER DIFFERENCES

ABSTRACTAs strongly suggested by recent work, patterns of gender difference can change because of changes in the selectivity of the sample itself. This is a statistical influence connected with the distributions of female and male scores, rather than a substantive influence related to demographic characteristics of the sample such as age or ethnicity. It is, nonetheless, an important influence because gender differences that are partly statistical in origin can easily confuse possible implications regarding education and assessment.This report proposes a general model to account for the effects of sample restriction on gender differences. Simulations showed the model to be quite accurate in reproducing standard mean differences and other statistics in a restricted sample. Three primary contributing factors were identified: the range‐restricting effects of sample selection, differential variability of female and male scores in the original sample, and the representation of females and males in the restricted sample. A test with actual data showed reasonably good consistency between trends predicted by the model and trends in gender differences that have been widely observed in advanced tests administered to select samples.

Maximum Likelihood Inference from Sample Survey Data

Summary In this paper we present a general theory for maximum likelihood inference based on sample survey data. Our purpose is to identify and emphasise the recurring basic concepts that arise in the application of likelihood methods, including the estimation of precision, to survey data. We discuss the problems generated by the effects of sample design, selection and response processes. We also discuss the problem of failures of the model assumptions and the role of sample inclusion probabilities in achieving robustness. We present two illustrative examples, one of which illustrates the use of non-Gaussian models.

Interpretation of negative sample selection effects in wage offer equations

The finding of negative sample selection effects in estimated wage offer equations is often viewed as being ‘problematic’, resulting from misspecification of the wage and selection processes. This note shows that negative sample selection effects are plausible in the context of Heckman's reservation wage model of labour supply. Therefore, such findings need not necessarily be cause for concern.

Work experience, earnings and participation: evidence from the women and employment survey

This paper presents a probit model of labour force participation and a set of earnings equations estimated using data from the 1980 Women and Employment Survey. The results of the participation model compare with previous findings and the earnings models give typical results of human capital and job characteristics variables. Exclusion of the sample selection effect appears to be of importance and a linear piecewise experience variable outperforms the quadratic specification.

IMMIGRANT WAGE DIFFERENTIALS AND THE ROLE OF SELF‐EMPLOYMENT IN AUSTRALIA*

digitalData = {"site":{"ip":"173.234.59.127","environment":"LIVE","website":"onlinelibrary.wiley.com","websiteCode":"pericles","serverDate":"2022-04-30"},"identities":[{"type":"SmartGroupUser","uuid":"b238ab85-6bb4-49e4-84ca-a0d9f3ee5a35"},{"type":"SmartGroupUser","uuid":"eabce89b-65c2-4a09-ab7c-bd2b3fea2f10"},{"type":"RegionUser","uuid":"3c5060a1-6620-438f-862b-902a4d4cd4ac"},{"type":"SmartGroupUser","uuid":"a86d27d2-6bcd-4b94-bca5-fe16f4882855"},{"type":"SmartGroupUser","uuid":"da88d808-3f7b-4bde-9210-10862b6d121b"},{"type":"SmartGroupUser","uuid":"8c72a733-117a-4bc9-a682-13475f13e6a3"},{"type":"SmartGroupUser","uuid":"756fc162-8722-461f-a077-ebffec4f4fae"},{"type":"SmartGroupUser","uuid":"cf9c85b9-7f3b-4309-aeaa-c41669cabe3b"},{"type":"SmartGroupUser","uuid":"db8b8d10-1aed-4809-b3ee-70938c569014"},{"type":"SmartGroupUser","uuid":"cdce41c2-a8de-4a64-a4d9-a85caaaa2314"},{"type":"SmartGroupUser","uuid":"7a2a316d-5a37-4ccd-afdb-737bfb863a48"},{"type":"SmartGroupUser","uuid":"39f911ce-44b6-4dba-9978-8b2e947a8eb0"},{"type":"SmartGroupUser","uuid":"97587ffd-dcb4-4cb5-9986-a8ce611532bd"},{"type":"SmartGroupUser","uuid":"a46a154d-979f-41ba-b459-fd34df5c071f"},{"type":"SmartGroupUser","uuid":"b7acda8c-ce18-451b-a68d-e14b72f9b5b8"},{"type":"SmartGroupUser","uuid":"055a2bdd-861f-4fa7-bd49-9773db7e0e3e"},{"type":"SmartGroupUser","uuid":"b2dbfe27-ecae-4bb6-bb0d-d75e5caa3762"},{"type":"SmartGroupUser","uuid":"1a6297fb-6679-4882-a4c2-fdd124624edf"},{"type":"SmartGroupUser","uuid":"8032f17e-5050-4fb4-9c40-e1f6875d5c2b"},{"type":"SmartGroupUser","uuid":"ab743440-628f-4964-9216-9983ad3beadf"},{"type":"SmartGroupUser","uuid":"a4187e00-f28e-4754-9a0f-1eb306fb0679"}],"publication":{"item":{"title":"IMMIGRANT WAGE DIFFERENTIALS AND THE ROLE OF SELF‐EMPLOYMENT IN AUSTRALIA<link href='#fn1'>*</link>","doi":"10.1111/j.1467-8454.1993.tb00133.x","access":"no","earliestDate":"1993-06-01","formats":["Cited by","First page image","PDF","References"],"topics":[{"taxonomyUri":"global-subject-codes","topicLabel":"Asia & Australasia","topicUri":"ec44"},{"taxonomyUri":"global-subject-codes","topicLabel":"General & Introductory Economics","topicUri":"ec00"},{"taxonomyUri":"global-subject-codes","topicLabel":"ECONOMICS","topicUri":"ec"},{"taxonomyUri":"global-subject-codes","topicLabel":"International Economics & Trade","topicUri":"ec40"}]},"group":{"doi":"10.1111/aepa.1993.32.issue-60","access":"no","volume":"32","issue":"60","coverDate":"1993-06-01","topics":[{"taxonomyUri":"global-subject-codes","topicLabel":"Asia & Australasia","topicUri":"ec44"},{"taxonomyUri":"global-subject-codes","topicLabel":"General & Introductory Economics","topicUri":"ec00"},{"taxonomyUri":"global-subject-codes","topicLabel":"ECONOMICS","topicUri":"ec"},{"taxonomyUri":"global-subject-codes","topicLabel":"International Economics & Trade","topicUri":"ec40"}]},"series":{"title":"Australian Economic Papers","doi":"10.1111/(ISSN)1467-8454","access":"no","code":"14678454","issn":"0004-900X","eissn":"1467-8454","topics":[{"taxonomyUri":"gpt-sitename","topicLabel":"wly.economics_000047","topicUri":"wly.economics_000047"},{"taxonomyUri":"global-subject-codes","topicLabel":"Asia & Australasia","topicUri":"ec44"},{"taxonomyUri":"global-subject-codes","topicLabel":"General & Introductory Economics","topicUri":"ec00"},{"taxonomyUri":"global-subject-codes","topicLabel":"ECONOMICS","topicUri":"ec"},{"taxonomyUri":"global-subject-codes","topicLabel":"International Economics & Trade","topicUri":"ec40"},{"taxonomyUri":"gpt-zonename","topicLabel":"austr-econ-papers_AEPA","topicUri":"austr-econ-papers_aepa"}]}},"page":{"subPage":"Abstract","page":"Article/Chapter View","pageGroup":"Publication Pages","title":"title","errorCode":200}}

The relative effectiveness of private and public schools: Evidence from two developing countries

We estimate quality differences between private and public secondary schools in two developing countries: Colombia and Tanzania. Quality is measured by student performance on standardized achievement tests. We focus on the sample selection aspects of making quality comparisons. Estimated sample selection effects suggest that Colombian student sort themselves by type of institution (private or public); but Tanzanian students appear to be selected by a hierarchical mechanism, with worse students entering private institutions. These effects are consistent with the different institutional frameworks for educational choice in these countries. For both countries, private schools offer an achievement advantage. Standardizing for differences in student and school attributes, private school students have higher achievement test scores.

High galactic latitude cataclysmic variables

A sample of stars which are either known or are good candidates for being halo cataclysmic variables has been isolated. This sample consists of 84 stars of which 57 are dwarf novae (two of these are probable DQ Her systems), 12 are novalike stars, seven are AM Her objects, and eight are classical novae. Periodicities are known for 27 of the systems. If these stars are indeed halo members, then their distances range from 350 pc to 8 kpc and they represent objects that should be significantly older than their disk counterparts. If they do not represent true halo members, then they are intrinsically very faint systems (Mv greater than +10). Comparing the gross properties of this sample to those for the disk cataclysmic variables, it is found that below the period gap, the mean outburst amplitude for the halo dwarf novae is 3 mag greater than for the disk dwarf novae. For systems with periods less than 4 h, the halo cataclysmic variables have a higher percentage of systems below the period gap than the disk cataclysmic variables. The implications of these findings are discussed, as well as possible sample biases and selection effects.

Interpreting sample selection effects

This note examines the structure of a two-sector selection model with a probit selector equation. The sector selected depends on each observation's characteristics, indexed by θ i , i − 1,…, N. Each sector has a distinct regression equation and the expected values of the dependent variable for these equations vary with the sector selected and, through the sample selection terms, θ i . The sample selection effect is interpreted by examining the differences in the expected values of the dependent variables in the sector equations as the sector selected and θ i change. The standard case where there is only one sector is studied as a special case.

Effects of sample selection on the coincidence of hypertension and diabetes

Effects of sample selection on the coincidence of hypertension and diabetes

Effects of Sample Selection on the Coincidence of Hypertension and Diabetes

The occurrence of both hypertension and diabetes in patients at a general medical clinic at a university center and within a work-site community has been determined. In addition, the course of such patients for four years following presentation has been observed. The combination of hypertension and diabetes appeared more frequently, was expressed by greater severity, and coincided with a higher observed morbidity among the hospital clinic patients than among the work-site population. These findings regarding the prevalence and natural history of these coexistent diseases reaffirm that the source of a selected population can significantly affect the pattern of disease expression observed and, therefore, limit the capacity to extrapolate results from one population to another.